Abstract
`A number of drugs are sensitive to the presence of oxygen,
`resulting in oxidative degradation and a shortened shelf-life.
`This is usually minimized by either the addition of antioxi(cid:173)
`dants to the formulation or the replacement of headspace air
`with an inert gas, such as nitrogen. A new alternative is pre(cid:173)
`sented and involves the use of an oxygen extractor.
`The purpose of this study was to demonstrate the effec(cid:173)
`tiveness of the device in removing oxygen from drug vials .
`Aqueous epinephrine solutions were prepared from deoxy(cid:173)
`genated water. Twenty milliliters of the solution was filled
`into vials to which caps were affixed; some caps contained the
`oxygen extractor and others did not. The oxygen content of
`the headspace was analyzed and the appearance of the solu(cid:173)
`tions was observed over a period of up to 19 days.
`The results showed that the vials with the oxygen extractor
`remained clear for the 19 days' duration with essentially no
`oxygen present; whereas the vials without the unit discolored
`within 24 hours, turning to black, oily films in 15 days. This
`technology may provide compounding pharmacists with a
`method of packaging oxygen-sensitive preparations with en(cid:173)
`hanced stability.
`
`Introduction
`No organic drug is insensitive to oxygen. However, many are
`oxidized at a sufficiently slow rate to be considered stable. Other
`drugs related to the family of catechols, benzamides, thiazines,
`tetracyclines, etc. are oxygen sensitive and have limited shelf(cid:173)
`life. Shelf-life extension is afforded by addition of oxygen scav(cid:173)
`engers (antioxidants) , which are basically more oxygen avid than
`the drug. More recently, preservation of these drugs also includes
`the improvement of drug-container closures and drug processing
`under an inert gas. H owever, in essence, antioxidants are also life
`limited and common processing practices still allow 1 % to 3 % of
`oxygen in the inert processing gas.
`Some extremely oxygen-sensitive drugs, which are essen tially
`oxygen intolerant, require elaborate processing conditions and
`equipment with a tolerance limit of 200 ppm of oxygen. It is need(cid:173)
`less to elaborate on the investment required to achieve such stan(cid:173)
`dards and on the need for oxygen-resistant storage containers for
`these drugs.
`Background
`The basic concept of this technology is to remove oxygen via an
`electrochemical process that consists of applying a voltage avail(cid:173)
`able from a hearing-aid battery, or a button cell, to an electro(cid:173)
`chemical cell; reducing oxygen available within the drug vial and
`releasing oxygen at a counterelectrode, thereby effectively trans(cid:173)
`ferring oxygen out .of the vial. The pra ctical considerations un(cid:173)
`derlying the implementation of such a concept are to provide: (a)
`
`PEER REVIEWED
`
`Use of an Oxygen Extractor
`to Minimize Oxidation
`of Compounded Preparations
`
`Henri J. R. Maget, PhD
`Med-E-Cell
`1063 3 Roselle Street, Suite E
`San Diego, CA 92121-1506
`
`Table 1. Extractor Useful Life.
`
`Battery Q-value
`(mAhour)
`
`Extractor Life Battery Size, mm
`(years)
`(Diameter/Height)
`
`50 - 60
`
`110 - 120
`
`210-220
`
`2
`
`4
`
`5.8/3.6
`
`7 .8/3 .6
`
`7.8/5 .3
`
`an economical solution to the prevention of oxidative degradation
`of drugs in vials, (b) an oxygen-free environment in vials without
`the need to change the vial configuration and geometry, and (c) a
`means for oxygen elimination by incorporating only minor mod(cid:173)
`ifications to the existing conventional vial elastomeric closures.
`The oxygen extractor has only two components, namely, a com(cid:173)
`mercial battery (button cell) selected for the specific purpose to
`be accomplished, generally 5 .8 to 11.6 mm in diameter and 3 .6
`to 5.3 mm high; and an electrochemical cell, which is a mem(cid:173)
`brane fitted with two electrodes, about 0.2 mm thick. Structurally,
`once assembled, these two components can be less than 10 mm
`in diameter and less than 6 mm in elevation. On standby, the ex(cid:173)
`tractor is inactive. It becomes activated by closing the circuit be(cid:173)
`tween the battery and the cell, which can be accomplished by a
`variety of mechanical means, at any time before, during or after
`vial closure .
`The most important variable affe cting the extractor perfor(cid:173)
`mance is the battery storage capacity. It will be used to extract oxy(cid:173)
`gen from the initial air headspace in the vial and to r emove the
`oxygen diffusing back into the vial, mainly through the elas(cid:173)
`tomeric closure (stopper).
`Miniature commercial batteries (button cells) have an oxygen
`extraction capacity of 10 to 50 cc ofoxygen. This capability is ad(cid:173)
`equate t o maintain oxygen-free environments for one to four
`years for most elastomeric closures used for drug vials .
`Oxygen-concentration levels can be achieved and maintained be(cid:173)
`low 200 ppm, representing a thousandfold concentration decrease
`over air-filled vials and a hundredfold decrease over inert-gas(cid:173)
`processed vials.
`Extractor Life Expectancy
`The life in hours of the extractor, T, can be estimated from the
`following correlation:
`- T = (6/R)(Q-V)
`where R is the rate (cc/ hour) of oxygen diffusion from the ambi (cid:173)
`ent air into the vial; Q is the battery storage capacity, in mil-
`
`International Journal of P harmaceuti cal Compounding
`Vo/.3 No.6 November/December 1999
`
`, •
`
`Eton Ex. 1028
`1 of 3
`
`

`

`PEER REVIEWED
`
`Table 2. Extractor Operating Time Required
`for a Twentyfold Reduction in Oxygen Concentration.
`
`Gas Phase
`Volume (cc)
`
`Extractor
`Characteristic K
`
`Time to Achieve
`1 % Oxygen from Air
`
`12
`
`0.7
`
`12
`
`50
`
`50
`
`121
`
`3.2 hours
`
`11.3 minutes
`
`1.3 hours
`
`Table 3. Appearance of Epinephrine
`Solutions, at 60°C, in Absence and Presence
`of the Oxygen Extractor.
`
`21
`20
`
`10
`
`z
`""' "' >
`X
`1 - -- - ---------------
`0
`~ oL-~-- -- - +-::=:==iL_- - ---c---1- -=:---~
`2.00
`2.50
`0.00
`0.50
`3.0
`3.50
`4.00
`0.19
`3.1
`
`TIME, HOURS
`
`Visual Appearance
`of the Reference
`Solution without
`Time
`(Hours) Extractor
`
`Visual Appearance
`of the Vial
`Measure
`with th e
`Oxygen Extractor Oxygen %
`
`Conditions: I. Silver-oxide battery, model 76; V = 21 cc; K = 50 _
`11. Silver-oxide battery, model 76: V = 21 cc; K = 212
`Ill. Zn-Air battery, model 675; V = 0.71 cc; K = 50
`
`0
`
`16
`
`45
`
`88
`
`135
`
`184
`
`280
`
`425
`
`456
`
`Clear
`
`Faint yellow
`
`Light yellow
`
`Clear
`
`Clear
`
`Clear
`
`Light yellow/brown
`
`Clear
`
`Golden brown
`
`Clear
`
`Dark golden brown
`
`Clear
`
`Dark amber brown
`
`Clear
`
`Oily film appears
`
`Clear
`
`Black with oil film
`
`Clear
`
`NM = not measured
`* The sensor is inaccurate below 0.1 %
`
`NM
`
`1.9
`
`0.4
`
`<0.1 *
`
`<0.1 *
`
`<0.1 *
`
`<0 .1 *
`
`NM
`
`NM
`
`liamphours; and Vis the initial gas-phase volume of the vial, cm3.
`For example, if the closure value R is 0.0 3 cc/day of oxygen, and
`V = 15 cc, the extractor's useful life can be estimated, as reported
`in Table 1, which also includes the size of the commercial batteries
`available to perform the extraction task.
`Theory
`Since the rate of oxygen r emoval from the gas phase is propor(cid:173)
`tional to the oxygen content, the oxygen pressure at any time t
`(hours) can be predicted from:
`Po2 = P 0 02 e-kt
`Equation 1
`where P 0 02 is the original oxygen pressure in the vial and k is a
`system constant that depends on the initial gas volume, V cc, and
`a characteristic K of the extractor, such that k = K/V The value(cid:173)
`of K depends principally on the active surface area of the extrac(cid:173)
`tor. Since the extractor current is also proportional to P02, the same
`type of relationship will hold for the battery current, namely:
`Equation 2
`I= I 0 e-kt
`where I 0 is the initial extractor current for P 0 02- These unique
`
`Internationa!Journal of Pharmaceutical Compounding
`Vol.3 No .6 Nove11tber/December 1999
`
`Fig . 1. Extractor removing oxygen from containers.
`
`correlations suggest that m easurements of the current becom e
`an indication of the level of oxygen depl etion.
`Equations 1 and 2 are useful in predicting the time required to
`achieve a desired oxygen concentration level. For example, the time
`required to reduce the oxygen pressure from 0.21 to 0.01 atm, (or
`from 0.03 to 0.0015 atm) will be obtained from
`t = (1/k) Ln 21 = 13 .2 (V/K), as illustra ted in Table 2
`These calculated values compare favorably with the experi(cid:173)
`mental results presented in F ig. 1.
`Experiment
`All experiments were conducted with an extractor assembly
`consisting of an electrochemical cell and a small battery. Electrical
`connections were mad e via a series-resistor placed between the
`battery anode and the electrochemical cell cathode. The extrac(cid:173)
`tor was attached to chambers sealed from the environment. In
`some instances, an oxygen sensor, (Microelectrodes, Inc., Bedford,
`NH) amperometric oxygen probe, mod el MI 73 0, and oxygen
`meter model OM-4 were placed in the gas chamber. The sensor
`is only accurate for oxygen concentrations above 0.1 % . This ex(cid:173)
`perimental setup could be used to monitor battery current and oxy(cid:173)
`gen concentration, simultaneously.
`Experiments I and II of the figure were performed with a sil(cid:173)
`ver-oxide button cell, model 7 6 (11.6 -mm diameter/5 .3 5-mm
`height), with the same gas-phase volume, V = 21 cc, but differ(cid:173)
`ent ch aracteristics, K, of the extractor. Experim ent III was
`performed with a model 675 zinc-air battery (11.6-mm diame(cid:173)
`ter/5 .3 5-mm height).
`The oxygen con centration decreases pre dictably. Below 0.1 % ,
`the current progressively decreases to a "mainte-nance" current
`corresponding to the rate of diffusion of ambient oxygen into the
`chamb er or vial. For a diffusion rate of 0.03 cc/day, the mainte(cid:173)
`nance current is approximately 5 µA.
`
`Eton Ex. 1028
`2 of 3
`
`

`

`Oxygen Extraction
`from Drug Vials
`
`Prototype extractors were used to deter(cid:173)
`mine their effectiveness in preventing ox(cid:173)
`id ative degradation of epinephrine, which
`is known to be sensitive t o oxygen. The
`rate of degradation was accentua ted by
`using concentrated aqueous solutions
`(1:1250) at elevated temper ature (60°C),
`in absence of any retardation agents, an(cid:173)
`ti oxidants or stabilizer s. Epin eph rine (as a
`bitartrate) solutions were prepared by us (cid:173)
`ing deoxygen ated water and the system was
`assem bled under a blanket of nitrogen.
`T he 38-mL glass vials were filled with 20
`mL of the solution, adjusted to a pH of 4.
`In the experiment reported in Table 3, the
`extractor included an oxygen probe, de(cid:173)
`scribed previously.
`Results
`The reference solutions, without extrac(cid:173)
`tors, displayed a coloration within 24
`hours, turning to black, oily films in 15
`days; whereas the vial fitt ed with the ex(cid:173)
`tractor remained clear throughout the test
`duration, namely, 19 days. These stabili ty
`results were confirmed by researchers at
`the U niversity of Tennessee by measuring
`the epinephrine concentration over a period
`of 24 days. 1
`Conclusion
`
`This technology allows removal of oxy(cid:173)
`gen and maintenance of an essentially oxy(cid:173)
`gen-free environment in containers, and is
`suitable for drug vials or any other appli(cid:173)
`cation requiring the creation and mainte(cid:173)
`nance of anaerobic environments. It can be
`applied to contained solids, liquids or so(cid:173)
`lutions, including a varie ty of oxygen(cid:173)
`sensitive products, such as cosmetics, food
`and beverages, chemicals, etc.
`
`References
`
`1. Wood G, Aksorntoae N, Thoma L.
`Evaluation of an oxyge n extraction
`device in slowing epinephrine degra(cid:173)
`dation. Presented at th e Parenteral
`Drug Associa tion Annual Meeting,
`Washington, DC, Fall 1998. (cid:127)
`
`PEER REVIEWED
`
`I ·N D E X of
`
`ADVERTISERS
`
`Analytical Research Laboratories . ... 482
`Apothecary Products .. . .... .. ... . ... 451
`Application Design Consultants ...... 446
`B & B Pharmaceuticals, Inc .......... 489
`Chemical and Pharmaceutical
`Industry Co., Inc .. .. ...... ... ....... 438
`Display Options, Inc ...... . .. ....... . 456
`Flavorx .. . . ..... ................... 455
`Gallipot, Inc ..... .. ..... ... 450, 487, OBC
`Gladson & Associates .. . . .. ......... 426
`ISO Tech Design . .. ........... . ..... 485
`Lorann Oils . ... ............... ... ... 486
`Mallinckrodt, Inc ...... ... ... ........ 466
`Medica Instrumentation ........... .. 483
`
`Meridian Chemical. .. . ........... .. . 460
`PK Software . . ...... . .. ... .. . ....... 489
`Paddock Laboratories, Inc .... . . . . .. . . 459
`Pharmaceutical Specialties, Inc ... ... 484
`Pharma-Tek, Inc ..... .. .... . .... . ... 442
`Professional Compounding
`Centers of America, Inc . . 436-437, 465, IFC
`Respiratory Distributors, Inc ... .. . 452-453
`Robert P. Potts & Assoc., Inc .. . . .... .. 495
`Rx Formulator ................. .. ... 502
`Spectrum Quality Products, Inc ....... IBC
`The Compounder's Website .. ....... . 492
`Triple Crown America ............ . .. 440
`
`For advertising information, contact LAUREN BERNICK.
`Address: P. 0. Box 890546 , Houston TX 77289 USA
`Tel: 800 -6 61-4572 Fax: 800-494-4572
`email: lbernick@ijpc.com
`
`STORE PLANNING
`STORE FIXTURES
`& STORE DECOR
`
`Potts & Associates is the firm with the experience to assist pharmacists in
`planning their IV Clean Room , Compounding Laboratory, and developing a
`more professional image for their stores. Call us today for more information.
`SERVING PHARMACISTS SINCE 1954
`
`a Ro&ellt p. potts & assoc.
`
`2722 Fairmont, Dallas, Texas 75201
`(214) 871-8747 • 800-255-5498 • fax: (214) 969-6937
`
`Internationa l Journal of Pharmaceutical Compounding 1•
`
`Vol.3 No.6 November/D ecember 1999
`
`Eton Ex. 1028
`3 of 3
`
`